Revisiting the Sibling Head in Object Detector
Open accessAbstract
The "shared head for classification and localization'' (sibling head), firstly denominated in Fast RCNN, has been leading the fashion of the object detection community in the past five years. This paper provides the observation that the spatial misalignment between the two object functions in the sibling head can considerably hurt the training process, but this misalignment can be resolved by a very simple operator called task-aware spatial disentanglement (TSD). Considering the classification and regression, TSD decouples them from the spatial dimension by generating two disentangled proposals for them, which are estimated by the shared proposal. This is inspired by the natural insight that for one instance, the features in some salient area may have rich information for classification while these around the boundary may be good at bounding box regression. Surprisingly, this simple design can boost all backbones and models on both MS COCO and Google OpenImage consistently by ~3% mAP. Further, we propose a progressive constraint to enlarge the performance margin between the disentangled and the shared proposals, and gain ~1% more mAP. We show the TSD breaks through the upper bound of nowadays single-model detector by a large margin (mAP 49.4 with ResNet-101, 51.2 with SENet154), and is the core model of our 1st place solution on the Google OpenImage Challenge 2019.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Jointly learning localization and classification may cause sub-optimal solutions for both tasks, and decoupling localization and classification is helpful.
“As pointed out by recent works [7, 43], jointly learning of localization and classification may cause sub-optimal solutions for both tasks, and it is helpful to decouple localization and classification.”
From Learning Spatio-Temporal Transformer for Visual Tracking (2021) · cites this paper as [43]
Cites
ImageNet Large Scale Visual Recognition Challenge
Russakovsky, O., Deng, J., Su, H., Krause, J., Satheesh, S., Ma, S., Huang, Z., Karpathy, A., Khosla, A., Bernstein, M., et al. (2015) · International Journal of Computer Vision · doi:10.1007/s11263-015-0816-y · Cited by 39,380 · see citing works
Feature Pyramid Networks for Object Detection
Lin, T.-Y., Dollar, P., Girshick, R., He, K., Hariharan, B., & Belongie, S. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.106 · Cited by 30,939 · see citing works
Focal Loss for Dense Object Detection
Lin, T.-Y., Goyal, P., Girshick, R., He, K., & Dollar, P. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.324 · Cited by 30,581 · see citing works
Girshick, R. (2015) · 2015 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.169 · Cited by 28,105 · see citing works
Deformable Convolutional Networks
Dai, J., Qi, H., Xiong, Y., Li, Y., Zhang, G., Hu, H., & Wei, Y. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.89 · Cited by 7,513 · see citing works
Cascade R-CNN: Delving Into High Quality Object Detection
Cai, Z., & Vasconcelos, N. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00644 · Cited by 7,370 · see citing works
FCOS: Fully Convolutional One-Stage Object Detection
Tian, Z., Shen, C., Chen, H., & He, T. (2019) · 2019 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00972 · Cited by 7,013 · see citing works
Stacked Hourglass Networks for Human Pose Estimation
Newell, A., Yang, K., & Deng, J. (2016) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-46484-8_29 · Cited by 5,634 · see citing works
Deformable ConvNets V2: More Deformable, Better Results
Zhu, X., Hu, H., Lin, S., & Dai, J. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2019.00953 · Cited by 2,943 · see citing works
Soft-NMS — Improving Object Detection with One Line of Code
Bodla, N., Singh, B., Chellappa, R., & Davis, L. S. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.593 · Cited by 2,274 · cited 1× · see citing works
“For comparing with the best single-model TridentNet, in TSD, we apply the same configuration with it including multi-scale training, soft-NMS [1], deformable convolutions and the 3× training scheme on ResNet-101.”
MethodBottom-Up Object Detection by Grouping Extreme and Center Points
Zhou, X., Zhuo, J., & Krahenbuhl, P. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2019.00094 · Cited by 1,088 · see citing works
MegDet: A Large Mini-Batch Object Detector
Peng, C., Xiao, T., Li, Z., Jiang, Y., Zhang, X., Jia, K., Yu, G., & Sun, J. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00647 · Cited by 311 · see citing works
Revisiting RCNN: On Awakening the Classification Power of Faster RCNN
Cheng, B., Wei, Y., Shi, H., Feris, R., Xiong, J., & Huang, T. (2018) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-01267-0_28 · Cited by 282 · see citing works
Hao, Z., Liu, Y., Qin, H., Yan, J., Li, X., & Hu, X. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.207 · Cited by 134 · see citing works
Self-citation
Flagged first, then by citations
Cited by
Sorted by citations
Deformable DETR: Deformable Transformers for End-to-End Object Detection
Zhu, X., Su, W., Lu, L., Li, B., Wang, X., & Dai, J. (2020) · arXiv · doi:10.48550/arxiv.2010.04159 · Cited by 1,837 · see citing works
“4 TSD (Song et al., 2020) SENet154 + DCN X 51.”
Self-citationTOOD: Task-aligned One-stage Object Detection
Feng, C., Zhong, Y., Gao, Y., Scott, M. R., & Huang, W. (2021) · 2021 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.00349 · Cited by 1,668 · see citing works
“design, the task-interactive features inevitably introduce a certa n level of feature conflicts between two different tasks, which have also been discussed in [26, 28].”
BackgroundLearning Spatio-Temporal Transformer for Visual Tracking
Yan, B., Peng, H., Fu, J., Wang, D., & Lu, H. (2021) · 2021 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.01028 · Cited by 1,280 · see citing works
“As pointed out by recent works [7, 43], jointly learning of localization and classification may cause sub-optimal solutions for both tasks, and it is helpful to decouple localization and classification.”
BackgroundConditional DETR for Fast Training Convergence
Meng, D., Chen, X., Fan, Z., Zeng, G., Li, H., Yuan, Y., Sun, L., & Wang, J. (2021) · 2021 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.00363 · Cited by 1,055 · see citing works
Object Detection with Transformers: A Review
Shehzadi, T., Hashmi, K. A., Liwicki, M., Stricker, D., & Afzal, M. Z. (2025) · Sensors · doi:10.3390/s25196025 · Cited by 93 · see citing works
Equipment Identification and Localization Method Based on Improved YOLOv5s Model for Production Line
Yu, M., Wan, Q., Tian, S., Hou, Y., Wang, Y., & Zhao, J. (2022) · Sensors · doi:10.3390/s222410011 · Cited by 10 · see citing works
- Authors
- Song, G., Liu, Y., & Wang, X.
- Affiliations
- Sensetime (China) (CN)
- Chinese University of Hong Kong (HK)
- Venue
- 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
- Publisher
- IEEE
- Published
- 2020-06
- Pages
- 11560-11569
- ISSN
- 1063-6919 (ISSN-L)
- OpenAlex
- W3035473155
- Cited by
- 509
- References
- 14
- Citations by year
- 2026: 26 · 2025: 72 · 2024: 98 · 2023: 136 · 2022: 95 · 2021: 65 · 2020: 17
- FWCI
- 24.14
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- Licence
- Locations
- IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Visual Attention and Saliency Detection
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Image and Video Retrieval Techniques
- Keywords
- object detection, spatial misalignment, classification and regression, bounding box regression, Fast R-CNN
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
